Germinal matrix hemorrhage (GMH) occurs in approximately 45% of very low birth weight neonates. Slightly more than 1% of live births in the U.S. weigh less that 1.5_kg, thus it can be estimated that GMH affects 10,000 neonates annually. These babies will account for 30% of patients who will later exhibit cerebral palsy. GMH will continue to be a perplexing medical problem due to technological advances allowing survival of very premature babies, and socioeconomic conditions associated with their births. Little is known concerning specific factors and mechanisms involved in the development of GMH. Recent discussion of the physiologic factors predisposing to GMH concern unstable arterial blood pressure, ischemic brain injury, and wide swings in venous pressure. Certain large, fragile germinal matrix (GM) periventricular vascular "channels" from which bleeding often occurs have been characterized as "veins," "capillary sinusoids," or "a vascular rete." Previous investigations of the vascular anatomy of the GM are flawed. In ordinary neuropathological techniques, the sections are too thin to trace vessels; furthermore, veins cannot be discriminated from arteries because these immature vessels do not have mural smooth muscle or collagen. Studies employing intravascular injection into cadaveric vasculature produce an unintelligible mass of vessels and artifacts. Errors in this basic concept may have adversely influenced strategies for the salvage of these babies. Clearly, innovative techniques are required to investigate the nature of vascular morphology and pathology in the neonate with GMH. The native alkaline phosphatase (AP) ectoenzyme, present in the endothelial plasma cell membrane of small arteries, arterioles, and capillaries, but not of venule/veins, can be exploited to differentially stain the GM vasculature in thick celloidin sections. This histochemical technique is superior to injection methods of outlining vasculature because it avoids artifacts of injection and the background tissue can be examined with counterstains. The broad objective of this study is to characterize the GM angioarchitecture, identify the precise site(s) of bleeding, and determine the ultimate fate of the immature GM sinusoids with AP staining and thick celloidin sections in the brains from 40 pediatric subjects ranging from 22- to 40-weeks' post-conception age.